Project Summary: CpG islands (CGIs) function as promoters for greater than 60% of human genes. Importantly, these elements remain mostly free of CpG methylation - an epigenetic mark associated with stable transcriptional silencing - despite the fact that the vast majority of CpG sites in the genome are methylated. While the relative immunity of CGIs against epigenetic silencing has been noted for years and is critical to their function, its underlying mechanism has remained elusive. Here, we report three novel and key observations relevant to this process: (i) a large fraction of CGI promoters, while GC-rich overall, display marked strand asymmetry, or skew, in the distribution of G and C residues; (ii) transcription through such regions of high GC-skew leads to the formation of long R-loop structures in which the newly transcribed G-rich RNA remains hybridized to the template C-rich DNA strand, forcing the non-template DNA strand into a largely single-stranded conformation; and (iii) R-loop formation protects the underlying DNA sequence from the action of DNA methyltransferases (DNMTs). Based on this knowledge, we hypothesize that R-loop formation at mammalian CGI promoters serves to protect these regions against epigenetic silencing. We propose to further test this hypothesis through three Specific Aims combining computational, genomics, biochemical, and molecular genetics approaches in human and mouse cells. Specific Aim 1: To test the hypothesis that R-loop formation is a widespread and conserved property of mammalian CGI promoters. Specific Aim 2: To test the hypothesis that R-loop formation protects against DNA methylation. Specific Aim 3: To test the hypothesis that altered R-loop formation leads to aberrant DNA methylation patterns.